The song in question not a genuine track, but a convincing fake created by "research and deployment company" OpenAI, whose Jukebox project uses artificial intelligence to generate music, complete with lyrics, in a variety of genres and artist styles. Along with Sinatra, they've done what are known as "deepfakes" of Katy Perry, Elvis, Simon and Garfunkel, 2Pac, Céline Dion and more. Having trained the model using 1.2m songs scraped from the web, complete with the corresponding lyrics and metadata, it can output raw audio several minutes long based on whatever you feed it. Input, say, Queen or Dolly Parton or Mozart, and you'll get an approximation out the other end.

The first issue of image analysis is classification. The problem is that we get a picture at the input and we want to get information at the output. The classification is defined by assigning an object to a group of objects called a class. It assumes that we have a group of objects and assign them to classes at the same time. Each object should be characterized by appropriate features: shape, color, proportions of different characteristic sizes. In order to assign an object to a particular class, it is necessary to find a set of features that can be found automatically in the image.

Reinforcement learning (RL) has recently shown impressive performance in complex game AI and robotics tasks. To a large extent, this is thanks to the availability of simulated environments such as OpenAI Gym, Atari Learning Environment, or Malmo which allow agents to learn complex tasks through interaction with virtual environments. While RL is also increasingly applied to natural language processing (NLP), there are no simulated textual environments available for researchers to apply and consistently benchmark RL on NLP tasks. With the work reported here, we therefore release NLPGym, an open-source Python toolkit that provides interactive textual environments for standard NLP tasks such as sequence tagging, multi-label classification, and question answering. We also present experimental results for 6 tasks using different RL algorithms which serve as baselines for further research. The toolkit is published at https://github.com/rajcscw/nlp-gym

Recent advances in commonsense reasoning depend on large-scale human-annotated training data to achieve peak performance. However, manual curation of training examples is expensive and has been shown to introduce annotation artifacts that neural models can readily exploit and overfit on. We investigate G-DAUG^C, a novel generative data augmentation method that aims to achieve more accurate and robust learning in the low-resource setting. Our approach generates synthetic examples using pretrained language models, and selects the most informative and diverse set of examples for data augmentation. In experiments with multiple commonsense reasoning benchmarks, G-DAUG^C consistently outperforms existing data augmentation methods based on back-translation, and establishes a new state-of-the-art on WinoGrande, CODAH, and CommonsenseQA. Further, in addition to improvements in in-distribution accuracy, G-DAUG^C-augmented training also enhances out-of-distribution generalization, showing greater robustness against adversarial or perturbed examples. Our analysis demonstrates that G-DAUG^C produces a diverse set of fluent training examples, and that its selection and training approaches are important for performance. Our findings encourage future research toward generative data augmentation to enhance both in-distribution learning and out-of-distribution generalization.

This paper proposes CANC, a Co-teaching Active Noise Cancellation method, applied in spatial computing to address deep learning trained with extreme noisy labels. Deep learning algorithms have been successful in spatial computing for land or building footprint recognition. However a lot of noise exists in ground truth labels due to how labels are collected in spatial computing and satellite imagery. Existing methods to deal with extreme label noise conduct clean sample selection and do not utilize the remaining samples. Such techniques can be wasteful due to the cost of data retrieval. Our proposed CANC algorithm not only conserves high-cost training samples but also provides active label correction to better improve robust deep learning with extreme noisy labels. We demonstrate the effectiveness of CANC for building footprint recognition for spatial computing.

Predictions obtained by, e.g., artificial neural networks have a high accuracy but humans often perceive the models as black boxes. Insights about the decision making are mostly opaque for humans. Particularly understanding the decision making in highly sensitive areas such as healthcare or fifinance, is of paramount importance. The decision-making behind the black boxes requires it to be more transparent, accountable, and understandable for humans. This survey paper provides essential definitions, an overview of the different principles and methodologies of explainable Supervised Machine Learning (SML). We conduct a state-of-the-art survey that reviews past and recent explainable SML approaches and classifies them according to the introduced definitions. Finally, we illustrate principles by means of an explanatory case study and discuss important future directions.

The process of association and tracking of sensor detections is a key element in providing situational awareness. When the targets in the scenario are dense and exhibit high maneuverability, Multi-Target Tracking (MTT) becomes a challenging task. The conventional techniques to solve such NP-hard combinatorial optimization problem involves multiple complex models and requires tedious tuning of parameters, failing to provide an acceptable performance within the computational constraints. This paper proposes a model free end-to-end approach for airborne target tracking system using sensor measurements, integrating all the key elements of multi target tracking -- association, prediction and filtering using deep learning with memory. The challenging task of association is performed using the Bi-Directional Long short-term memory (LSTM) whereas filtering and prediction are done using LSTM models. The proposed modular blocks can be independently trained and used in multitude of tracking applications including non co-operative (e.g., radar) and co-operative sensors (e.g., AIS, IFF, ADS-B). Such modular blocks also enhances the interpretability of the deep learning application. It is shown that performance of the proposed technique outperforms conventional state of the art technique Joint Probabilistic Data Association with Interacting Multiple Model (JPDA-IMM) filter.

In this paper, we propose a rule-based engine composed of high quality and interpretable regular expressions for medical text classification. The regular expressions are auto generated by a constructive heuristic method and optimized using a Pool-based Simulated Annealing (PSA) approach. Although existing Deep Neural Network (DNN) methods present high quality performance in most Natural Language Processing (NLP) applications, the solutions are regarded as uninterpretable black boxes to humans. Therefore, rule-based methods are often introduced when interpretable solutions are needed, especially in the medical field. However, the construction of regular expressions can be extremely labor-intensive for large data sets. This research aims to reduce the manual efforts while maintaining high-quality solutions

In the management of computer network systems, the service function chaining (SFC) modules play an important role by generating efficient paths for network traffic through physical servers with virtualized network functions (VNF). To provide the highest quality of services, the SFC module should generate a valid path quickly even in various network topology situations including dynamic VNF resources, various requests, and changes of topologies. The previous supervised learning method demonstrated that the network features can be represented by graph neural networks (GNNs) for the SFC task. However, the performance was limited to only the fixed topology with labeled data. In this paper, we apply reinforcement learning methods for training models on various network topologies with unlabeled data. In the experiments, compared to the previous supervised learning method, the proposed methods demonstrated remarkable flexibility in new topologies without re-designing and re-training, while preserving a similar level of performance.